In the Scientific American April 2012 issue there is a piece talking about the notions of information density vs number of syllables per second as according to Peter Roach's observation:
Speakers of some languages seem to rattle away at high speed like machine-guns, while others sound rather slow and plodding. (1998)
The Scientific American piece itself is quite interesting if a bit light and uninformative for those conversant in linguistics.
As a person who grew up in a territory that ostensibly has two writing systems (syllabics and "roman orthography") and as a thinker interested in the technical issues of language instruction, I've spent some time thinking about the pros and cons of either writing system for pedagogical purposes.
The syllabic writing system is a phonetic system (ie, one syllable, one symbol) as opposed to the "roman orthography" (which is phonemic, like English or French, and denotes segments rather than syllables). The strength of the syllabic system is that it is compact and easy to master (even for non-Inuktitut speakers) because it relies upon orientation to denote the attendant vowel (upright for 'i'; right for 'u'; left for 'a'); but it is realtively difficult to impart sight-reading skills, and to discern recurrent patterns (therefore, difficult to standardize spelling conventions) which is important for grammar learning rather than script learning.
The "roman orthography" which is based on latin script like English is easier to discern recurrent patterns, and, in fact, very analytic-friendly (ie, most linguists use the latin script to analyse grammatical features of Inuktitut). The problem with latin script rendering of Inuktitut is that words or phrases can get pretty long. This is not a problem in itself because the important patterns are easier to see in latin script than in syllabics and sight-reading becomes possible, but the problem here has more to do with aesthetics.
Just for fun, I started one day sitting in a plane waiting to take off to count how many symbols are used in English and French signs on average. No more than seven or eight letters on average are used in both English and French signs; Inuktitut phrases in latin script average around ten letters. Some phrases can go up to 15 or more letters.
But this is not as problematic is first assumed. Most Inuktitut phrases contain recurrent morphemes, and most morphemes tend to occur in regular patterns (ie, morphemes occur in clusters and in an ordered way - like, [-nngit-] the negative morpheme tends to occur after verbs (am, are, is) and tenses (past, present and future) and right before pronominal endings.
The Scientific American piece says:
In the 1950s linguist Noam Chomsky proposed the idea of universal grammar, which suggests that all languages, their apparent differences notwithstanding, possess a common set of of abstract structures. This hypothesis galvanized the field of linguistics, but truly common structures proved tough to find. (p. 18, Scientific American April 2012)
The last sentence above is not strictly truthful. The abstract "structures" common to all languages are the lexical classes: verbs, nouns, their modifiers (adjectives and adverbials). Every natural language structure possess these different classes that along with specific, and highly rigid and rigorous, grammatical rules (ie, cases and moods) make up all human languages.
At the -emic level, even mathematics may be embraced as a natural language though granted instead of nouns and verbs numeral values (variables and constants) act as its elements. Mathematics, in this regard, is less about arithmetic operations than syntactical rules that denote relations and orders in which things occur.
The Dirac equations on the electron, for eg, is said to be an equation that just keeps on giving. Its informational density is immense and rich beyond its austere appearance. Like poetry it goes deeper than its "words"; unlike poetry, nothing is frivolously interpreted and created by the interpreter. The logical structure of the Dirac equation is deep and profound and subtle.